The present invention relates to a method and an apparatus for analyzing and synthesizing human speech.
A speech analysis and synthesis device, which uses an analysis and synthesis method, such as LPC, PARCOR, LSP, or the Formant method, is well known. In order to compress human speech information, the device stores a quantization parameter as information representing human speech. The quantization parameter is generated by quantizing a feature parameter which is extracted from human speech. The device synthesizes human speech on the basis of the quantization parameter,
In conventional devices, the feature parameter is linearly quantized so that the quantization parameter is proportional to the feature parameter. When the quantization parameter comprises N bits, the value of the quantization parameter ranges from 0 to 2.sup.N-1. Sometimes the distribution of the feature parameter is localized in the center of the range. However, it is not desirable to reduce the number N of bits of the quantization parameter. This is because the reduction of the number N of bits causes the range of the quantization parameter to be narrowed. This results in the problem that the quantization parameter has unnecessary redundancy.
In order to solve the above-mentioned problem, another conventional device quantizes a transformation parameter which is generated by nonlinearly transforming a feature parameter. The feature parameter can be represented by means of the quantization parameter having the necessary minimum number of bits. However, when nonlinearly quantizing, it is necessary to reversely and nonlinearily transform the quantization parameter to generate a synthesis parameter. This reverse and nonlinear transformation is extremely complicated. This results in the problem that the speech analysis and synthesis device requires very large circuits.